BACKGROUND: Hypercortisolism is known to cause osteoporosis. OBJECTIVE: To evaluate the prevalence of subclinical hypercortisolism in participants referred for evaluation of osteoporosis. DESIGN: Cross-sectional study. SETTING: Two community hospitals and research institutes in Italy. PATIENTS: 219 patients without clinically overt hypercortisolism or other secondary causes of osteoporosis who were referred for evaluation of osteoporosis between January 2005 and December 2005. MEASUREMENTS: Bone mineral density was measured by using dual-energy x-ray absorptiometry, and hypercortisolism was assessed with serum cortisol levels after a dexamethasone suppression test. Also measured were 24-hour urinary free cortisol levels and midnight plasma cortisol levels. RESULTS: Seven of 65 patients with T-scores of 2.5 or less and vertebral fractures had subclinical hypercortisolism (prevalence, 10.8% [95% CI, 3.23% to 18.31%]). This prevalence was 4.8% (CI, 1.32% to 8.20%) among patients with osteoporosis. In multivariable analyses adjusted for age, sex, and body mass index, a positive dexamethasone suppression test result was associated with the presence of osteoporosis (odds ratio, 3.37 [CI, 1.78 to 6.43]; P < 0.001) and vertebral fractures (odds ratio, 1.70 [CI, 1.04 to 2.79]; P = 0.035). LIMITATIONS: The study was conducted in a referral setting; its findings may not apply to the general population. CONCLUSIONS: Subclinical hypercortisolism may be more common than is generally recognized in patients with osteoporosis in whom secondary causes of osteoporosis have been excluded.